Genetics of Behçet's disease: lessons learned from genomewide association studies.

REVIEW URRENT C OPINION

Genetics of Behçet’s disease: lessons learned from genomewide association studies Ahmet Gu¨l

Purpose of review Behçet’s disease is a complex disease, and genetic susceptibility plays a critical role. This review aimed to discuss the recent genomewide association study (GWAS) findings and their implications to the pathogenesis of Behçet’s disease. Recent findings GWAS data confirmed the major role of HLA-B51 in Behçet’s disease susceptibility, and the discovery of epistatic interactions between HLA-B51 and ERAP1 variants provided some hints about its possible pathogenic mechanisms. Investigation of human leukocyte antigen (HLA) Class I region showed weaker but independent associations around HLA-A and HLA-C regions. Genomewide studies also established associations with IL10, IL23R, CCR1, STAT4, KLRC4, GIMAP2/GIMAP4, and UBAC2 genes in Behçet’s disease patients of different ethnicities. Deep resequencing of targeted genes identified additional associations with rare variants in TLR4, MEFV, and NOD2 genes. Summary GWAS data established a major step forward by providing insights into the underlying mechanisms in Behçet’s disease with the discovery of new susceptibility genes. These variations may implicate defects in the sensing and processing of microbial and endogenous danger signals as well as in the regulation of innate and adaptive immune responses in Behçet’s disease. Association findings with HLA Class I antigens as well as IL23R, ERAP1, IL10, and MEFV genes also suggest shared inflammatory pathways with spondyloarthropathies. Keywords Behçet’s disease, ERAP1, HLA-B51, IL-10, IL-23R, MHC Class I

INTRODUCTION Behçet’s disease is a multisystem inflammatory disorder characterized by recurrent exacerbations affecting mucocutaneous tissues, eyes, blood vessels, and several other tissues. It has a multifactorial cause, and a complex genetic background plays an important role in the susceptibility to Behçet’s disease. Familial aggregation with a relatively high sibling recurrence risk ratio (ls value between 11.4 and 52.5) and identification of a strong association with HLA-B51 have been accepted as the main clues documenting the genetic contribution to its pathogenesis [1 ]. The first evidence of the human leukocyte antigen (HLA) Class I association (HL-A5) with Behçet’s disease was reported in 1973 by the analysis of 21 Japanese patients with Behçet’s disease and 78 healthy controls [2], and the risk allele was later defined as HLA-B51 [1 ,3]. It is still the strongest association finding described so far, and it has been

replicated in several different ethnic groups. Metaanalysis of available HLA studies conducted in Behçet’s disease patients of different ethnicities revealed a considerably high odds ratio (OR) of 5.78 for HLA-B5/B51 allele for the development of Behçet’s disease [4]. On the other hand, case–control studies using candidate-gene approach were not as successful in the identification of non-HLA genes, and only a few of the reported findings were replicated in different ethnic groups. Replication problems of

&

&

www.co-rheumatology.com

Division of Rheumatology, Department of Internal Medicine, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey Correspondence to Ahmet Gu¨l, Division of Rheumatology, Department of Internal Medicine, Istanbul Faculty of Medicine, Istanbul University, 34093 Fatih, Istanbul, Turkey. Tel: +90 212 631 8699; e-mail: dr.agul [email protected]; [email protected] Curr Opin Rheumatol 2014, 26:56–63 DOI:10.1097/BOR.0000000000000003 Volume 26 Number 1 January 2014

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Genetics of Behçet’s disease Gu¨l

KEY POINTS

additional data on the remaining parts of HLA Class I region and several non-HLA genes [6–8,9 –11 ]. Therefore, this article aimed to discuss the recent GWAS findings and their implications to the pathogenesis of Behçet’s disease. &&

HLA-B51 is the strongest susceptibility factor for Behçet’s disease, but pathogenic mechanisms related to HLA-B51 remain to be elucidated. Epistatic interaction between HLA-B51 and ERAP1 implicates a direct pathogenic role for HLA-B51 molecule, by affecting the peptide repertoire, their binding affinity, and proper folding of heavy chain. Other polymorphisms within the HLA Class I located in the vicinity of HLA-B and MICA genes as well as HLA-C and HLA-A regions may also contribute to the disease pathogenesis. Identified non-HLA variations may implicate defects in the sensing and processing of microbial and endogenous danger signals as well as in the regulation of innate and adaptive immune responses in Behçet’s disease pathogenesis. Association with IL10, IL23R, ERAP1, and MEFV genes suggest shared inflammatory pathways with spondyloarthritides.

the association findings were mainly because of either the lack of enough statistical power or not selecting the right candidates because of the absence of reliable data on the disease pathogenesis [5]. Genomewide association study (GWAS) approach has gained importance as a valuable tool for the analysis of common polymorphisms in the pathogenesis of complex disorders. During the last 3 years, the results of six GWASs conducted in Turkish, Japanese, Chinese, and Korean patients with Behçet’s disease were reported. Those studies both confirmed the association of Behçet’s disease with the HLA-B locus and also revealed important

&&

GENOMEWIDE ASSOCIATION STUDIES GWAS has been developed for the investigation of common (minor allele frequency 1%) singlenucleotide polymorphisms (SNPs) to the disease phenotype by using different low-cost, highthroughput genotyping platforms. Six different GWAS data are now available for Behçet’s disease. As outlined in Table 1, there are important differences between these studies, such as the sample sizes, patient characteristics, replication samples, genotyping platforms, number of SNPs, and use of stringent genomewide significance levels. However, despite all their limitations, replicated GWAS findings have provided invaluable input to the genetics of Behçet’s disease and understanding of the pathogenic pathways.

HLA CLASS I REGION AND BEHC ¸ ET’S DISEASE The major histocompatibility complex (MHC) on chromosome 6p21 contains HLA and other immune response-related genes. HLA Class I region provided the strongest association findings with Behçet’s disease in all GWAS groups, with an exception of the study by Fei et al. [6], which reported no MHC region association finding achieving genomewide significance. A detailed analysis of the Turkish and Japanese data revealed quite similar patterns in the HLA Class

Table 1. Summary of the genomewide association studies in Behçet’s disease Fei et al. [6]

Remmers et al. [7]

Mizuki et al. [8]

Hou et al. && [9 ]

Lee et al. && [10 ]

Kirino et al. && [11 ]

Turkish

Turkish

Japanese

Chinese

Korean

Turkish

Patients

152

1215

612

149

379

1209

Controls

Ethnicity Discovery samples

172

1278

740

951

800

1278

Replication samples

No

Japanese, Turkish, Middle Eastern Arab, Greek, Korean, British

Turkish, Korean

Chinese

Japanese

Turkish, Japanese

Genomewide significance

Permutation P value

5 108

5 108

5 108

False discovery rate

5 108

Number of analyzed SNPs

500 Ka

311 459

320 438

661 736

594 591

779 465

SNP, single-nucleotide polymorphism. a No data are available for the number of analyzed number of SNPs in pooled samples.

1040-8711 ß 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins


57


Vasculitis syndromes

I region with an accumulation of strong association findings around the HLA-B gene and a weaker but independent association signals around the HLA-A/ HLA-F regions at the telomeric end of Class I [7,8]. For further analysis of the HLA-B region, Remmers et al. genotyped 1190 patients and 1257 controls for HLA with two-digit resolution, and their investigation revealed that HLA-B51 allele is more strongly associated with Behçet’s disease than any of the genotyped polymorphisms in the region (OR ¼ 3.49, 95% confidence interval 2.95–4.12). As linkage disequilibrium spans very long distances in the MHC region, HLA-B51 allele was identified on an exclusive extended haplotype, which was observed in 35.2% of the patients and 15.9% of the controls. However, the frequency of the same haplotype lacking only HLA-B51 was found to be the same (4%) in both groups. On the other hand, Hughes et al. recently reported their genotyping results of 8572 SNPs in the HLA region by ImmunoChip platform [12,13 ], conducted in two independent cohorts of Turkish (503 patients and 504 controls) and Italian (144 patients and 1270 controls) origin [14 ]. Using the data from the 1000 Genomes Project, they imputed additional variants in the HLA region reaching 33 054 variants in Turkish and 32 819 in Italian samples. Their analysis revealed association with three independent SNPs in the HLA region reaching a genomewide significance level, and the strongest finding was observed for rs116799036 (OR ¼ 3.88), which is located approximately 24 kb upstream of HLA-B and 18 kb upstream of MICA genes. Hughes et al. also used imputed their data for HLA-A, HLA-B, HLA-C, HLA-DQA1, HLA-DQB1, and HLA-DRB1 alleles. Pairwise conditional analysis using imputed HLA-B51:01 typing, as the major but not the only subtype of HLA-B51, showed rs116799036 remains significant independently of HLA-B51:01 in Turkish and Italian cohorts. Also, HLA-B51:01 association was found to disappear when the data were controlled for rs116799036, suggesting that the association of HLA-B51 with Behçet’s disease could be explained by rs116799036, and there may be no causal relationship with HLAB51 allele. This study provided no functional analysis for rs116799036 polymorphism regarding its possible effects either for HLA-B or MICA expression, very similar to the observations on the HLA-C expression in psoriasis and HIV because of variations in the enhancer regions [15,16]. Another recent study provided support for direct involvement of HLA-B51 molecule to the genetic tendency to Behçet’s disease. Re-evaluation of Turkish GWAS collection with the imputed genotypes of autosomal SNPs showed that homozygosity &

&&

58


for the ERAP1 gene rs17482078 variant encoding p.Arg725Gln preferentially conferred risk for HLAB51-positive Behçet’s disease patients with an OR of 3.78 [11 ]. A recessive association with ERAP1 polymorphism was observed only in HLA-B51 carriers, very similarly to the ERAP1 association with HLA-C06 in psoriasis and with HLA-B27 in ankylosing spondylitis patients, but with an opposite allelic effect in Behçet’s disease [17,18]. ERAP1 gene encodes an endoplasmic reticulum amino peptidase, which trims proteasome-derived peptides from their amino terminals before being loaded onto the antigen-binding groove of HLA Class I molecules. B pocket of the groove has a very important role in determining the peptide-binding affinity and disease susceptibility, and the size and polarity of this pocket affect the preferential binding of certain peptides depending mainly on the amino acids in the second position [1 ]. Therefore, editing of the amino terminals of peptides by different ERAP1 isoforms may change the peptide repertoire and their binding affinities, which may then affect the folding properties of HLA molecules as well as immunological synapse with CD8þ T cells [19]. In a region with a very strong linkage disequilibrium, functional contribution of rs116799036 and other polymorphisms in the vicinity of HLA-B and MICA loci remains to be elucidated. However, identification of a functional ERAP1 association only in HLA-B51-positive Behçet’s disease patients could be accepted as strong evidence for the involvement of HLA-B51 molecule directly in the pathogenesis of Behçet’s disease. The exact role of HLA-B51 in Behçet’s disease susceptibility has yet to be clarified, and it is highly likely that its contribution may involve more than one mechanism [1 ]. A recent meta-analysis investigating the association of HLA-B51/B5 allele with clinical manifestations of Behçet’s disease showed that its presence predominates in male patients, and HLAB51 positivity is associated with moderately higher prevalence of genital ulcers, ocular and skin manifestations, and a decreased prevalence of gastrointestinal involvement [20 ]. Available data indicate that HLA-B51 is not useful as a biomarker for the diagnostic or prognostic classification of the patients with Behçet’s disease. Turkish and Japanese GWAS data also showed independent association findings for the telomeric end of MHC Class I around the HLA-A region [7,8,14 ]. A previously detailed HLA genotyping in Japanese Behçet’s disease patients documented a weak association of HLA-A26:01, a common HLAA allele in the Japanese population, with Behçet’s disease (OR ¼ 1.92) [21]. The same study [21] also revealed weak associations with HLA-F01:01:01 and &&

&

&

&

&&

Volume 26 Number 1 January 2014



HLA-G01:01:02 as well as protective associations with HLA-A33:03, HLA-F01:01:02, and HLAG01:04:01, similarly to the previously reported protective associations with HLA-E01:01 and HLA-G01:01:01 in Korean Behçet’s disease patients [22]. A strong association of HLA-A26:01 was also observed in Korean Behçet’s disease patients with posterior uveitis (91.7%) compared to uveitis patients without posterior segment involvement (26.9%) [23]. Hughes et al.’s [14 ] ImmunoChip data showed an association with rs114854070 (OR ¼ 1.96), which tags HLA-F antisense RNA 1 (HLA-F-AS1) gene. Analysis of the same cohort using imputed HLA genotypes also revealed that HLA-A26:01 was not associated with Behçet’s disease in Turkish and Italian patients, but an independent association with HLA-A02:01 was observed in both cohorts [14 ]. Kurata et al. [24] previously reported independent associations with the nearby TRIM39 and RNF39 genes in this region as well. Differently from Remmers et al.’s and Mizuki et al.’s data, Hughes et al. [14 ] reported a third area of independent association in the HLA-C region, both with PSORS1C1 (rs12525170, OR ¼ 3.01) and HLA-Cw16. Previous linkage studies in familial Behçet’s disease cases revealed a broad linkage peak in the MHC region extending to the telomere [25], and most recent findings document several independent associations with Behçet’s disease for HLA alleles and other MHC genes (Fig. 1). Considering the strong linkage disequilibrium in the MHC region, &&

&&

we still need more detailed analyses in larger number of patients as well as functional investigations for having a robust idea about the independent roles of different MHC genes in Behçet’s disease susceptibility.

NON-HLA SUSCEPTIBILITY GENES IN BEHC ¸ ET’S DISEASE Family studies suggested the importance of nonHLA genes to the genetic susceptibility for Behçet’s disease [26]. However, candidate-gene approach and underpowered case–control studies were not very successful in the discovery of genuine non-HLA genetic associations. Therefore, GWAS approach was a milestone in the genetics of Behçet’s disease by providing strong evidence for association at least for eight genes, and a targeted investigation of rare variants added two more to the list of susceptibility genes for Behçet’s disease.

&&

IL10 and IL23R-IL12RB2 Initially, two main GWASs revealed genomewide significant association findings for IL10 and IL23R-IL12RB2 genes [7,8]. Remmers and colleagues identified an association with an intronic variation, rs1518111 (P ¼ 1.88 108), and Mizuki and colleagues with promoter region variations of rs1800872 and rs1800872 (P ¼ 9.5 108). In addition to cross-validation of the findings in Turkish and Japanese patients, IL10 association

HLA-B*51 HLA-A*02:01 HLA-A*26:01 HLA-A*33:03**

HLA-Cw16

HLA-A

Telomere

HLA-F HLA-G RNF39

HLA-C

HLA-B

MICA

HLA-E TRIM39

Centromere

PSORS1C1 MICA*009

HLA-F*01:01:01 HLA-F*01:01:02**

HLA-G*01:01:02 HLA-G*01:04:02** HLA-G*01:01:01**

rs116799036

HLA-E*01:01**

FIGURE 1. MHC Class I region associations in Behçet’s disease. Data from [7,8,14 ,21,22,24]. MHC, major histocompatibility complex. &&



59



was replicated in Middle Eastern Arab, Greek, British, and Korean Behçet’s disease patients. IL10 gene encodes interleukin (IL)-10, and disease-associated A-allele of rs1518111 variation was found to be associated with a reduced mRNA expression in monocytes (35% of the expression level of G-allele) by measuring the allelic imbalance in heterozygous individuals. Also, peripheral blood mononuclear cells or monocytes produced significantly less IL-10 following stimulation with Toll-like receptor (TLR) ligands, such as lipopolysaccharide or Pam3Cys þ muramyl dipeptide, respectively, in individuals homozygous for A-allele of rs1518111 [7]. Mizuki and colleagues identified an association with rs12119179 (P ¼ 2.7 108), which is located in the intergenic region between IL23R and IL12RB2 genes. In the Turkish cohort, no genomewide significant result was observed for IL23R-IL12RB2, but initial analysis showed three SNPs with P value less than 104. Fine mapping of the region identified rs924080 (P ¼ 5.35 106), located similarly in the intergenic region between IL23R and IL12B2 genes; and meta-analysis of the Turkish and Japanese data for rs924080 reached genomewide significance (P ¼ 6.69 109). Analysis of the associated SNPs and linkage disequilibrium blocks suggested that they were more likely to be associated with IL23R gene, but not with IL1RB2 block [7]. Although association with IL23R gene was cross-validated in Turkish and Japanese patients, it could not be replicated in Korean, Middle Eastern Arab, Greek, and British Behçet’s disease samples. Later, an independent group replicated the association of IL10 and IL23R genes with Behçet’s disease in an Iranian cohort with additional findings suggesting a role for regulatory regions of IL23R but not IL12RB2 gene for Behçet’s disease susceptibility [27 ]. Also, rs17375018 variation of IL23R gene and a haplotype of four SNPs were first reported to be strongly associated in a GWAS in Chinese patients [28]; however, no significant association finding for IL23R gene was later published among full GWAS data by the same group [9 ]. No functional data were available for the IL23R variations identified in these GWASs; however, targeted resequencing of the IL23R gene in Japanese and Turkish patients with Behçet’s disease revealed several new association findings including the significantly decreased frequency of those rare missense variations with a protective effect by reduced IL-23-dependent IL-17 production, similar to the findings in Crohn’s disease [29 ,30]. &

&&

&&

re-analysis of 1209 patients with Behçet’s disease and 1278 controls by increasing the number of investigated SNPs with the imputation method. Behçet’s disease-associated rs7616215 variant of CCR1 gene (C–C motif chemokine receptor 1) was found to be associated with the expression, by conferring disease risk causing lower expression in those carrying the risk allele, which results in reduced monocytes chemotaxis in response to CCR1 ligand MIP1-a. A meta-analysis using a group of independent Turkish cohort and Japanese samples confirmed the association with CCR1 [11 ]. Association of CCR variants with Behçet’s disease was also replicated by another group in Chinese patients [31 ]. Functional association with CCR1 gene and possible monocyte chemotaxis defect provided a new pathogenic insight into the Behçet’s disease pathogenesis by suggesting an insufficiency in the clearance of at least some microbial pathogens or in other monocyte functions [11 ]. Imputation analysis also showed suggestive association findings for STAT4, KLRC4, and IL12A genes, and those genes were further analyzed by direct genotyping of the original set, an independent Turkish cohort and Japanese samples. Association with the STAT4 gene rs7574070 variant was confirmed by meta-analysis using all three groups, and STAT4 expression was found to be higher in individuals carrying Behçet’s disease-associated risk allele [11 ]. Hou et al. [9 ] identified genomewide significant association findings in 10 genes along with 13 other genes including STAT4, with suggestive (P < 1 104) findings in the discovery set of a GWAS conducted in 147 Han Chinese patients with Behçet’s disease and 951 controls. During the replication phase using 554 patients and 1159 controls of Han Chinese, only STAT4 variants showed significant association findings, and rs897200 and rs7572482 variations of STAT4 reached the genomewide significance by the analysis of the combined set. Similarly, Hou and colleagues [9 ] found that Behçet’s disease-associated A allele of rs897200 variation was associated with higher STAT4 expression, a more severe disease course, and also interestingly with increased production of IL-17, despite association of STAT4 mainly with Th1 polarization. Suggestive protective association finding with rs2617170 (p.Asn104Ser) variation of KLRC4 (killer cell lectin-like receptor subfamily C, member 4) gene were confirmed by replication in Turkish and Japanese patients as well as combined analysis (OR ¼ 0.78, P ¼ 1.34 109) [11 ]. KLRC4 is located on a haplotype block, which harbors four other natural killer cell receptor genes including KLRK1, which encodes NKG2D, and this region overlaps &&

&

&&

&&

&&

&&

&&

CCR1, STAT4, and KLRC4 Kirino and colleagues identified additional associations with CCR1, STAT4, and KLRC4 genes with the 60





with the strongest peak described in the whole genome linkage analysis of familial Behçet’s disease cases [32]. Functional role of KLRC4 has yet to be elucidated, but the haplotype carrying the protective rs2617170 variation was reported to be associated with reduced peripheral blood cytotoxicity [33], which further suggests a possible pathogenic role for killer cell receptor associated interactions through HLA or MICA molecules in the pathogenesis of Behçet’s disease. On the other hand, IL12A variation (rs17810546) was not polymorphic in the Japanese population, and combined analysis of Turkish discovery and replication samples did not give a genomewide significant result.

ERAP1 Association with ERAP1 gene was identified during an analysis with recessive model using the imputation data. Its epistatic interaction with HLA-B51 and its pathogenic significance are discussed above.

Turkish patients was not found to be significant in Italian patients (meta-analysis OR ¼ 1.84, P ¼ 1.69 107). Sawalha et al. [34] reported that the Behçet’s disease-associated risk allele of rs7999348 was associated with increased expression of UBAC2. A recent study by Hou et al. [35 ] investigated all five GWAS findings of Fei et al. in Chinese population, and they confirmed the association of only UBAC2 gene with Behçet’s disease. However, the risk allele for Chinese Behçet’s disease patients (rs3825427, OR ¼ 1.5, P ¼ 6.9 106 for the combined set) was found to be associated with decreased expression of UBAC2 in the peripheral blood and skin [35 ]. The function of UBAC2 has not yet been clarified, but its ubiquitin-associated domain suggests a role in ubiquitination and protein degradation pathways. Association findings for another ubiquitin-pathway gene SUMO4 (small ubiquitinlike modifier 4) using candidate-gene approach in Chinese, Korean, and Tunisian Behçet’s disease patients may also support ubiquitin-associated mechanisms in the disease pathogenesis [36–38]. &

&

GIMAP &&

Lee et al. [10 ] performed a GWAS in a Korean cohort of 379 Behçet’s disease patients and 800 controls, and they found an association with a haplotype containing six GIMAP (GTPase of immunity-associated protein family) genes. Their association findings with GIMAP2 and GIMAP4 variations were replicated in a Japanese cohort of 363 patients and 272 controls. Lower expression of GIMAP1 and GIMAP4 in CD4þ, and GIMAP2 in CD8þ T cells was found in the peripheral blood of Behçet’s disease patients. However, no correlation was detected between genetic variations and expression levels [10 ]. &&

UBAC2 Fei et al. [6] reported the association findings for KIAA1529, CPVL, LOC100129342, UBASH3B, and UBAC2 genes with a GWAS conducted in a cohort of 152 Turkish Behçet’s disease patients and 172 healthy controls, though none of the polymorphisms reached genomewide significance. The same group did a replication study [34] later for UBAC2 (UBA domain containing 2) gene by further genotyping additional 376 patients and 369 controls of Turkish, and 144 patients and 560 controls of Italian origin. They replicated the association of an intronic functional polymorphism (rs7999348) in Turkish and Italian sets (meta-analysis OR ¼ 1.39, P ¼ 1.85 105). Strong association found with another UBAC2 polymorphism (rs9517668) in

MISSING HERITABILITY IN BEHC ¸ ET’S DISEASE Available GWASs in Behçet’s disease have their own limitations, and their data could not explain all heritability in Behçet’s disease. Analyses of much larger cohorts of phenotypically well defined patients and matched controls are necessary for the discovery of weaker associations and delineation of genetic contribution to the development of Behçet’s disease subsets, such as uveitis, vascular involvement, or papulopustular lesions/arthritis clusters [39]. Missing heritability can also partially be explained by rare or population-specific variants contributing to the Behçet’s disease susceptibility. Kirino et al. [29 ] recently reported the association of rare nonsynonymous variants in the TLR4, MEFV, and NOD2 genes with Behçet’s disease. They performed a deep exonic resquencing in 11 genes involved in innate immunity (IL1B, IL1R1, IL1RN, NLRP3, MEFV, TNFRSF1A, PSTPIP1, CASP1, PYCARD, NOD2, and TLR4) as well as 10 genes discovered by GWAS. They identified rare and low-frequency variants in IL23R, TLR4, and NOD2 genes associated with Behçet’s disease. As an example of population-specific rare variants, p.Met694Val of MEFV gene, causing autosomal recessively inherited familial Mediterranean fever, conferred a genomewide significant conferred Behçet’s disease risk in the Turkish population (OR ¼ 2.65; P ¼ 1.8 1012), but it could not be replicated in the Japanese set.


&&


61



Sensing and processing of pathogen and danger associated signals

Defective regulation of innate and adaptive immune response

IL-23R

CCR1 IL-10

STAT4

IL12A

Microbes TLR4

Innate immunity Neutrophils, monocytes, macrophages, γδ T cells, NK cells

MEFV NOD2 Endogenous danger signals Trauma Uric acid ?

UBAC2 Adaptive immunity CD8+ T cells, Th1 / Th17 polarization

HLA-B51 ERAP1 KLRC4

FIGURE 2. Pathogenic implications of GWAS findings in Behçet’s disease. GWAS, genomewide association studies.

CONCLUSION GWAS approach established a major step forward by providing insights into the underlying pathogenic mechanisms in Behçet’s disease with the discovery of new genes. The major role of HLA-B51 was confirmed in Behçet’s disease susceptibility, and discovery of epistatic interactions between HLAB51 and ERAP1 variants provided some hints about its possible pathogenic mechanisms. Investigation of HLA Class I region showed weaker but independent associations around HLA-A and HLA-C regions, which may also have a critical importance in the understanding of HLA-related Behçet’s disease risk. Identified non-HLA variations may implicate defects in the sensing and processing of microbial and endogenous danger signals as well as in the regulation of innate and adaptive immune responses in Behçet’s disease pathogenesis (Fig. 2). Also, in addition to strong HLA Class I antigen associations with Behçet’s disease, new findings with IL23R, ERAP1, IL10, and MEFV genes suggest shared inflammatory pathways with spondyloarthritides [18,40–44]. Despite the clinical and genetic similarities, pleiotropic effects of genetic variations and disease-specific HLA associations may help to explain phenotypic differences [45 ]. Therefore, analysis of larger collections including well defined subsets of patients is expected to identify Behçet’s &

62


disease-related pathogenic pathways further for the development of more targeted therapies and biomarkers [5]. Acknowledgements None. Conflicts of interest Disclosure: No funding was received for this work. There are no conflicts of interest.

REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest && of outstanding interest 1. Gul A, Ohno S. HLA-B51 and Behcet disease. Ocul Immunol Inflamm 2012; & 20:37–43. A comprehensive review of the possible HLA-B51-related pathogenic mechanisms. 2. Ono S, Aoki K, Sugiura S, et al. Letter: HL-A5 and Behcet’s disease. Lancet 1973; 2:1383–1384. 3. Ohno S, Ohguchi M, Hirose S, et al. Close association of HLA-Bw51 with Behcet’s disease. Arch Ophthalmol 1982; 100:1455–1458. 4. De Menthon M, Lavalley MP, Maldini C, et al. HLA-B51/B5 and the risk of Behcet’s disease: a systematic review and meta-analysis of case–control genetic association studies. Arthritis Rheum 2009; 61:1287–1296. 5. Gul A. Genome-wide association studies in Behcet’s disease: expectations and promises. Clin Exp Rheumatol 2011; 29 (4 Suppl. 67):S3–S5. 6. Fei Y, Webb R, Cobb BL, et al. Identification of novel genetic susceptibility loci for Behcet’s disease using a genome-wide association study. Arthritis Res Ther 2009; 11:R66.



Genetics of Behçet’s disease Gu¨l 7. Remmers EF, Cosan F, Kirino Y, et al. Genome-wide association study identifies variants in the MHC class I, IL10, and IL23R-IL12RB2 regions associated with Behcet’s disease. Nat Genet 2010; 42:698–702. 8. Mizuki N, Meguro A, Ota M, et al. Genome-wide association studies identify IL23R-IL12RB2 and IL10 as Behcet’s disease susceptibility loci. Nat Genet 2010; 42:703–706. 9. Hou S, Yang Z, Du L, et al. Identification of a susceptibility locus in STAT4 for && Behcet‘s disease in Han Chinese in a genome-wide association study. Arthritis Rheum 2012; 64:4104–4113. A GWAS conducted in Chinese patients and confirmed the STAT4 association. 10. Lee YJ, Horie Y, Wallace GR, et al. Genome-wide association study identifies && GIMAP as a novel susceptibility locus for Behcet’s disease. Ann Rheum Dis 2013; 72:1510–1516. A GWAS conducted in Korean patients and discovered the GIMAP associations. 11. Kirino Y, Bertsias G, Ishigatsubo Y, et al. Genome-wide association analysis && identifies new susceptibility loci for Behcet’s disease and epistasis between HLA-B51 and ERAP1. Nat Genet 2013; 45:202–207. The first imputation analysis showing the epistatic interaction of ERAP1 variations with HLA-B51 and identifying the CCR1, STAT4, and KLRC4 associations with Behçet’s disease. 12. Trynka G, Hunt KA, Bockett NA, et al. Dense genotyping identifies and localizes multiple common and rare variant association signals in celiac disease. Nat Genet 2011; 43:1193–1201. 13. Ricano-Ponce I, Wijmenga C. Mapping of immune-mediated disease genes. & Annu Rev Genomics Hum Genet 2013; 14:325–353. A comprehensive review about the mapping of common immune-mediated disease genes. 14. Hughes T, Coit P, Adler A, et al. Identification of multiple independent && susceptibility loci in the HLA region in Behcet’s disease. Nat Genet 2013; 45:319–324. This study provides the ImmunoChip data for the HLA region in Turkish and Italian patients. 15. Clop A, Bertoni A, Spain SL, et al. An in-depth characterization of the major psoriasis susceptibility locus identifies candidate susceptibility alleles within an HLA-C enhancer element. PLoS One 2013; 8:e71690. 16. Apps R, Qi Y, Carlson JM, et al. Influence of HLA-C expression level on HIV control. Science 2013; 340:87–91. 17. Strange A, Capon F, Spencer CC, et al. A genome-wide association study identifies new psoriasis susceptibility loci and an interaction between HLA-C and ERAP1. Nat Genet 2010; 42:985–990. 18. Evans DM, Spencer CC, Pointon JJ, et al. Interaction between ERAP1 and HLA-B27 in ankylosing spondylitis implicates peptide handling in the mechanism for HLA-B27 in disease susceptibility. Nat Genet 2011; 43:761–767. 19. Reeves E, Edwards CJ, Elliott T, et al. Naturally occurring ERAP1 haplotypes encode functionally distinct alleles with fine substrate specificity. J Immunol 2013; 191:35–43. 20. Maldini C, Lavalley MP, Cheminant M, et al. Relationships of HLA-B51 or B5 & genotype with Behcet’s disease clinical characteristics: systematic review and meta-analyses of observational studies. Rheumatology (Oxford) 2012; 51:887–900. A detailed analysis of the HLA-B51 association with the disease manifestations. 21. Meguro A, Inoko H, Ota M, et al. Genetics of Behcet disease inside and outside the MHC. Ann Rheum Dis 2010; 69:747–754. 22. Park KS, Park JS, Nam JH, et al. HLA-E0101 and HLA-G010101 reduce the risk of Behcet’s disease. Tissue Antigens 2007; 69:139–144. 23. Kang EH, Park JW, Park C, et al. Genetic and nongenetic factors affecting the visual outcome of ocular Behcet’s disease. Hum Immunol 2013; 74:1363– 1367. 24. Kurata R, Nakaoka H, Tajima A, et al. TRIM39 and RNF39 are associated with Behcet’s disease independently of HLA-B 51 and -A 26. Biochem Biophys Res Commun 2010; 401:533–537. 25. Gul A, Hajeer AH, Worthington J, et al. Linkage mapping of a novel susceptibility locus for Behcet’s disease to chromosome 6p22-23. Arthritis Rheum 2001; 44:2693–2696.

26. Gul A, Hajeer AH, Worthington J, et al. Evidence for linkage of the HLA-B locus in Behcet’s disease, obtained using the transmission disequilibrium test. Arthritis Rheum 2001; 44:239–240. 27. Xavier JM, Shahram F, Davatchi F, et al. Association study of IL10 and IL23R& IL12RB2 in Iranian patients with Behcet’s disease. Arthritis Rheum 2012; 64:2761–2772. A confirmation of the IL10 and IL23R associations in Iranian patients. 28. Jiang Z, Yang P, Hou S, et al. IL-23R gene confers susceptibility to Behcet’s disease in a Chinese Han population. Ann Rheum Dis 2010; 69:1325– 1328. 29. Kirino Y, Zhou Q, Ishigatsubo Y, et al. Targeted resequencing implicates the && familial Mediterranean fever gene MEFV and the toll-like receptor 4 gene TLR4 in Behcet disease. Proc Natl Acad Sci USA 2013; 110:8134–8139. The first investigation for rare variants and reveals associations with the TLR4, NOD2, and MEFV Met694Val variation. 30. Momozawa Y, Mni M, Nakamura K, et al. Resequencing of positional candidates identifies low frequency IL23R coding variants protecting against inflammatory bowel disease. Nat Genet 2011; 43:43–47. 31. Hou S, Xiao X, Li F, et al. Two-stage association study in Chinese Han & identifies two independent associations in CCR1/CCR3 locus as candidate for Behcet’s disease susceptibility. Hum Genet 2012; 131:1841–1850. A confirmation of the CCR1/CCR3 association in Chinese patients. 32. Karasneh J, Gul A, Ollier WE, et al. Whole-genome screening for susceptibility genes in multicase families with Behcet’s disease. Arthritis Rheum 2005; 52:1836–1842. 33. Hayashi T, Imai K, Morishita Y, et al. Identification of the NKG2D haplotypes associated with natural cytotoxic activity of peripheral blood lymphocytes and cancer immunosurveillance. Cancer Res 2006; 66:563–570. 34. Sawalha AH, Hughes T, Nadig A, et al. A putative functional variant within the UBAC2 gene is associated with increased risk of Behcet’s disease. Arthritis Rheum 2011; 63:3607–3612. 35. Hou S, Shu Q, Jiang Z, et al. Replication study confirms the association & between UBAC2 and Behcet’s disease in two independent Chinese sets of patients and controls. Arthritis Res Ther 2012; 14:R70. A confirmation of the UBAC2 association in Chinese patients. 36. Hou S, Yang P, Du L, et al. SUMO4 gene polymorphisms in Chinese Han patients with Behcet’s disease. Clin Immunol 2008; 129:170–175. 37. Park G, Kim HS, Choe JY, et al. SUMO4 C438T polymorphism is associated with papulopustular skin lesion in Korean patients with Behcet’s disease. Rheumatol Int 2012; 32:3031–3037. 38. Kamoun M, Ben Dhifallah I, Karray E, et al. Association of small ubiquitin-like modifier 4 (SUMO4) polymorphisms in a Tunisian population with Behcet’s disease. Clin Exp Rheumatol 2010; 28 (4 Suppl. 60):S45–S49. 39. Karaca M, Hatemi G, Sut N, et al. The papulopustular lesion/arthritis cluster of Behcet’s syndrome also clusters in families. Rheumatology (Oxford) 2012; 51:1053–1060. 40. Reveille JD, Sims AM, Danoy P, et al. Genome-wide association study of ankylosing spondylitis identifies non-MHC susceptibility loci. Nat Genet 2010; 42:123–127. 41. Franke A, McGovern DP, Barrett JC, et al. Genome-wide meta-analysis increases to 71 the number of confirmed Crohn’s disease susceptibility loci. Nat Genet 2010; 42:1118–1125. 42. Anderson CA, Boucher G, Lees CW, et al. Meta-analysis identifies 29 additional ulcerative colitis risk loci, increasing the number of confirmed associations to 47. Nat Genet 2011; 43:246–252. 43. Tsoi LC, Spain SL, Knight J, et al. Identification of 15 new psoriasis susceptibility loci highlights the role of innate immunity. Nat Genet 2012; 44:1341– 1348. 44. Cosan F, Ustek D, Oku B, et al. Association of familial Mediterranean feverrelated MEFV variations with ankylosing spondylitis. Arthritis Rheum 2010; 62:3232–3236. 45. Solovieff N, Cotsapas C, Lee PH, et al. Pleiotropy in complex traits: challenges & and strategies. Nat Rev Genet 2013; 14:483–495. A comprehensive review discussing the pleiotropic effects of genetic variations.



63


Genetics in glioma: lessons learned from genome-wide association studies.

Alcohol Dependence Genetics: Lessons Learned From Genome-Wide Association Studies (GWAS) and Post-GWAS Analyses.

Lessons learned from gene identification studies in Mendelian epilepsy disorders.

Lessons Learned about Neurodegeneration from Microglia and Monocyte Depletion Studies.

Regulation of magnesium balance: lessons learned from human genetic disease.

Genetics of Sickle Cell-Associated Cardiovascular Disease: An Expert Review with Lessons Learned in Africa.

Alzheimer's Disease: Lessons Learned from Amyloidocentric Clinical Trials.

Lessons learned from hospice care.

Genome-wide association studies of refractive error and myopia, lessons learned, and implications for the future.

An efficient empirical Bayes method for genomewide association studies.

Multiple comparison procedures for neuroimaging genomewide association studies.

Complement regulators in human disease: lessons from modern genetics.

Guidelines for Large-Scale Sequence-Based Complex Trait Association Studies: Lessons Learned from the NHLBI Exome Sequencing Project.

Lessons learned from genetic testing for channelopathies.

Maternal microchimerism: lessons learned from murine models.

Genetics of autoimmune diseases: perspectives from genome-wide association studies.

Lessons Learned.

From microRNA target validation to therapy: lessons learned from studies on BDNF.

Lessons learned from the brute beast.

Lessons learned from a SIRT2-selective inhibitor.

Lessons learned from litigation: maintaining professional boundaries.

Transradial access: lessons learned from cardiology.

From lessons learned to emerging practices.

Lessons learned from hospital Ebola preparation.